Convergent synthesis of adenophostin A analogues via a base replacement strategy

Article abstract of DOI:10.1039/a909347h

The first totally synthetic base-modified analogues of the natural product and potent D-myo-inositol 1,4,5-trisphosphate receptor agonist adenophostin A were efficiently synthesised from D-xylose and D-glucose using methodology employing base and surrogate base addition to a common disaccharide intermediate.

Full text of DOI:10.1039/a909347h

Convergent synthesis of adenophostin A analogues via a base replacement
strategy
Rachel D. Marwood, Satoshi Shuto, David J. Jenkins and Barry V. L. Potter*
Wolfson Laboratory of Medicinal Chemistry, Department of Pharmacy and Pharmacology, University of Bath,
Claverton Down, Bath, UK BA2 7AY. E-mail: b.v.l.potter@bath.ac.uk
Received (in Cambridge, UK) 26th November 1999, Accepted 5th January 2000
The first totally synthetic base-modified analogues of the
natural product and potent
D-myo-inositol 1,4,5-trisphos-
phate receptor agonist adenophostin A were efficiently
synthesised from -xylose and -glucose using methodology
D
D
employing base and surrogate base addition to a common
disaccharide intermediate.
D-myo-Inositol 1,4,5-trisphosphate [Ins(1,4,5)P3, 1] is a second
messenger which mediates the release of intracellular Ca2+ on
binding to its own ligand-gated receptor.1 The synthesis of
Ins(1,4,5)P3 analogues has provided insights into the structural
requirements for binding and Ca2+ release at the Ins(1,4,5)P3
receptor,2 although inositol-based mimics with a potency
exceeding that of Ins(1,4,5)P3 have yet to be synthesised.
Adenophostins A and B (2 and 3)3 were isolated from broths
of Penicillium brevicompactum, and were found to be 10- to
100-fold more potent than Ins(1,4,5)P3 in releasing Ca2+ and in
receptor binding assays.4,5 Design of chiral ligands based upon
carbohydrates rather than cyclitols could provide novel syn-
thetic signal transduction modulators. The biological activity of
minimal structural analogues synthesised so far has not
exceeded that of Ins(1,4,5)P35–11 and it is likely that the adenine
base is crucial for potency. Adenophostin A is finding
widespread use as a tool to investigate cell signalling mecha-
nisms12 and there have been three syntheses of the mole-
cule.13–15 We required base-modified analogues of 2 and report
here an efficient convergent route to five such compounds via a
common disaccharide intermediate.
The disaccharide required for Vorbru¨ggen condensations,
Scheme 1 Reagents and conditions: i, Bu2SnO, BnBr, Bu4NBr, MeCN, 4 Å
sieves, Soxhlet, reflux, 24 h, 78%; ii, PhCH(OMe)2, TsOH, DMF, 70 °C,
–MeOH, 2 h, 89%; iii, NaCNBH3–HCl, THF–Et2O, 3 Å sieves, room temp.,
5 min, 68%; iv, (a) Ac2O–DMSO, 18 h, room temp., (b) NaBH4, EtOH–
H2O (5+4), room temp., 1 h, 50% over two steps; v, Ac2O, pyridine, room
temp., 20 h, 100%; vi, PdCl2, MeOH–CH2Cl2 (1+1), room temp., 3 h, 78%;
vii, (MeO)2PNEt2, 1H-tetrazole, CH2Cl2, room temp., 30 min; viii,
dioxane–toluene (3+1), 4 Å sieves, room temp., 2 h then ZnCl2, AgClO4,
dark, room temp., 20 h, 81% (based on 8); ix, AcOH–H2O–(CH2OH)2
(14+6+3), reflux, 15 min, 75%; x, Ac2O, pyridine, room temp., 20 h, 82%;
xi, (a) purine, (Me3Si)2NH–Me3SiCl (2+1), reflux, 20 h, (b) TMSOTf,
(CH2Cl)2, reflux, 7 h, 61%; xii, conc. aq. NH3–MeOH (1+5), sealed vessel,
room temp., 20 h, 97%; xiii, (BnO)2PNPri2, 1H-tetrazole, CH2Cl2, room
temp., 30 min, then MCPBA, 278 °C, 5 min, 80%; xiv, wet Pd(OH)2/C,
MeOH–cyclohexene–H2O (11+5+1), reflux, 17 h, 56%.
1,2,3A,4A-tetra-O-acetyl-2A,5,6A-tri-O-benzyl-3-O-a-
D
-glucopyr-
anosyl-
D
-ribofuranose (4), was prepared as follows. Species 516
(Scheme 1) was selectively 5-O-benzylated by two different
methods. Regioselective ring opening of a 3,5-O-dibutyl-
stannylene derivative with BnBr yielded the 5-O-benzyl ether 6
(mp 45–47 °C; [a]D +2.9) in one step. A 78% yield of
benzylated products was isolated containing 90% of the desired
regioisomer 6. This method has since been reported by another
group.17 Alternatively, acid-catalysed treatment of 5 with
benzaldehyde dimethyl acetal gave the crystalline benzylidene
acetal 7 (mp 125 °C; [a]D +3.8) as a single diastereoisomer in
89% yield; regiospecific cleavage of this acetal with
NaCNBH3–HCl also gave 6 in 68% yield. Inversion of the
3-hydroxy group to give the acceptor 8 (mp 83–85 °C; lit.18
81–83 °C) was achieved by oxidation with Ac2O and DMSO
and reduction of the corresponding 3-ulose with NaBH4.17
+83.0). Phosphitylation15 gave the phosphite donor 12 as an oil
1
in a 1+1 anomeric mixture as judged by H and 31P NMR
Acetylation of 9, available in two steps from
D
-glucose,6 gave
spectroscopy.
diacetate 10 quantitatively and cleavage of the allyl glycoside
was achieved using PdCl2 to give 11 (mp 71–74 °C; [a]D
Donor 12 and acceptor 8 were coupled together in a similar
fashion to that described11b to give the desired disaccharide 13
19
Chem. Commun., 2000, 219–220
This journal is © The Royal Society of Chemistry 2000
219

(mp 125–127 °C; [a]D +101.6) in 81% yield; the reaction time
had to be extended because of the deactivating effects20 of the
donor 3,4-di-O-acetate protecting groups. Note the high
regiospecificity of this glycosylation, the a-coupled anomer
being the sole isolated product in high yield. Cleavage of the
isopropylidene acetal was accomplished by heating 13 at reflux
in aqueous AcOH containing ethylene glycol14 to give 14 (mp
126–128 °C; [a]D +116.5) which was acetylated to give 4.†
To exemplify our route the condensation of 4 with purine is
described, using the method of Vorbru¨ggen et al.21 with
Notes and references
† Selected data for 4: mp 105–107 °C (EtOAc–hexane); Rf 0.29 (EtOAc–
hexane 3+7); [a]2D0 +98.2 (c 0.2, CHCl3) (Found: C, 64.00; H, 6.15. Calc. for
C40H46O14: C, 63.98; H, 6.18%); dH(400 MHz; CDCl3) 1.86, 1.87, 1.93,
1.96 (12H, 4s, CH3CO), 3.29, 3.36 (2H, ABX, 2JAB 10.7, 3JAX 3.9, 3JBX 2.4,
H-6AA, H-6AB), 3.56 (1H, J 9.8, 3.4, H-2A), 3.63, 3.72 (2H, ABX, 2JAB 11.2,
3JAX 3.9, 3JBX 2.9, H-5A, H-5B), 3.88 (1H, ddd, J 10.3, 2.9, H-5A), 4.29 (1H,
AB, JAB 12.2, OCHHAr), 4.37–4.40 (1H, m, H-4), 4.47–4.57 (4H, m, 2 3
OCH2Ar), 4.63–4.64 (2H, m, H-3, OCHHAr), 5.03–5.08 (2H, m, H-1A, H-
4A), 5.33 (1H, d, J 4.9, H-2), 5.38 (1H, dd, J 9.3, H-3A), 6.12 (1H, s, H-1),
7.23–7.34 (15H, m, ArCH); dC (100.4 MHz; CDCl3) 20.45, 20.56, 20.63,
20.74, 20.83, 20.94, 21.05, 21.16 (8q, 4 3 CH3COa and b), 67.52 (t, C-6A),
68.85 (d, C-5A), 69.00, 69.28 (2t, C-5a and b), 71.89, 72.02 (2d, C-3Aa and b),
73.02 (d, C-2), 73.17, 73.26, 73.35, 73.41, 73.48, 73.55 (4t and 2d, C-
3a and b, OCH2Ara and b), 76.59, 76.76 (2d, C-2Aa and b), 81.27 (d, C-4),
96.28, 96.47 (2d, C-1Aa and b), 98.46, 98.50 (d, C-1a and b), 127.32, 127.61,
127.74, 127.91, 128.02, 128.36, 128.46 (7d, ArCH), 137.51, 137.76, 138.11
(3s, 3 3 C-1 of Bn rings), 169.35, 169.69, 170.17, 170.28 (4s, 4 3 CH3CO);
a and b subscripts denote signals arising from a and b-anomers
respectively; m/z (FAB+) 750 (M+, 1%), 91 (100). For uridophostin: dH
(D2O) 3.90–3.50 (7H, m, H-4A, H-5A, H-2B, H-5B, and H-6B), 4.17 (1H, m,
H-4B), 4.32–4.28 (2H, m, H-3A and H-3B), 4.74 (1H, m, H-2A), 5.15 (1H, br
s, H-1B), 5.74 (1H, d, J 8.3, H-5), 5.95 (1H, d, J 4.4, H-1A), 7.65 (1H, d, J
8.3, H-6); dP(100 MHz, D2O, 1H-decoupled) 0.23, 1.09, 1.78; HRMS
(triethylammonium salt, FAB) calc. for C15H24N2O20P3 645.0135, found
645.0130 (100%, M2); lmax(H2O)/nm 260.
TMSOTf as catalyst. The major product was the 9-b- -
D
ribofuranosidopurine nucleoside (nebularine) derivative 15
([a]D +73.6) which exhibited a deshielded doublet at dH 6.44 (J
4.9 Hz) corresponding to H-1A of a b-substituted product; purine
signals in the 13C NMR spectrum also corresponded closely to
those of the known22 2A,3A,5A-tri-O-acetyl nebularine. Stirring 15
in a mixture of concentrated aqueous ammonia and methanol
gave triol 16 ([a]D +23.5) required for phosphorylation. Triol
16 was phosphitylated and the resulting trisphosphite was
oxidised to trisphosphate 17 ([a]D +17.9).15 Deprotection of 17
to the purine analogue of 2 (‘purinophostin’, 18) was achieved
with catalytic transfer hydrogenation. The free acid was eluted
from MP1 AG ion exchange resin with a gradient of aqueous
TFA and converted into the sodium salt.
Other adenine-related base surrogates used were benzimida-
zole and imidazole, condensation being achieved similarly to
above to give 19 and 20. The imidazole analogue of adenophos-
tin A (‘imidophostin’) 20, in which the adenine six-membered
1 M. J. Berridge, Nature (London), 1993, 361, 315.
2 B. V. L. Potter and D. Lampe, Angew. Chem., Int. Ed. Engl., 1995, 34,
1933.
3 S. Takahashi, T. Kinoshita and M. Takahashi, J. Antibiot., 1994, 47,
95.
4 M. Takahashi, K. Tanzawa and S. Takahashi, J. Biol. Chem., 1994, 269,
369; J. Hirota, T. Michikawa, A. Miyawaki, M. Takahashi, K. Tanzawa,
I. Okura, T. Furuichi and K. Mikoshiba, FEBS Lett., 1995, 368, 248.
5 J. S. Marchant, M. D. Beecroft, A. M. Riley, D. J. Jenkins, R. D.
Marwood, C. W. Taylor and B. V. L. Potter, Biochemistry, 1997, 36,
12780 and references therein.
6 D. J. Jenkins and B. V. L. Potter, Carbohydr. Res., 1996, 287, 169.
7 R. A. Wilcox, C. Erneux, W. U. Primrose, R. Gigg and S. R. Nahorski,
Mol. Pharmacol., 1995, 47, 1204.
8 N. Moitessier, F. Chre´tien, Y. Chapleur and C. Humeau, Tetrahedron
Lett., 1995, 46, 8023.
9 C. T. Murphy, A. M. Riley, C. J. Lindley, D. J. Jenkins, J. Westwick and
B. V. L. Potter, Mol. Pharmacol., 1997, 52, 741.
10 D. J. Jenkins, R. D. Marwood and B. V. L. Potter, Chem. Commun.,
1997, 449; Corrigendum, 805.
11 (a) S. Shuto, K. Tatani, Y. Ueno and A. Matsuda, J. Org. Chem., 1998,
63, 8815; (b) R. D. Marwood, A. M. Riley, V. Correa, C. W. Taylor and
B. V. L. Potter, Bioorg. Med. Chem. Lett., 1999, 9, 453.
12 Y. Huang, M. Takahashi, K. Tanzawa and J. W. Putney Jr., J. Biol.
Chem., 1998, 273, 31815; G. S. Bird, M. Takahashi, K. Tanzawa and J.
W. Putney Jr., J. Biol. Chem., 1999, 274, 20643; L. M. Broad, D. L.
Armstrong and J. W. Putney Jr., J. Biol. Chem., 1999, 274, 32881 and
references therein.
13 H. Hotoda, M. Takahashi, K. Tanzawa, S. Takahashi and M. Kaneko,
Tetrahedron Lett., 1995, 36, 5037.
14 N. C. R. van Straten, G. A. van der Marel and J. H. van Boom,
Tetrahedron, 1997, 53, 6509.
15 R. D. Marwood, V. Correa, C. W. Taylor and B. V. L. Potter,
Tetrahedron: Asymmetry, 2000, in press.
16 J. Moravcova, J. Capkova and J. Stanek, Carbohydr. Res., 1994, 263,
61.
17 T. Q. Chen and M. M. Greenberg, Tetrahedron Lett., 1998, 39, 1103;
P. B. Alper, M. Hendrix, P. Sears and C.-H. Wong, J. Am. Chem. Soc.,
1998, 120, 1965.
18 T. Desai, J. Gigg and R. Gigg, Carbohydr. Res., 1996, 280, 209.
19 W. Liao and D. Lu, Carbohydr. Res., 1996, 296, 171.
20 N. L. Douglas, S. V. Ley, U. Lu¨cking and S. L. Warriner, J. Chem. Soc.,
Perkin Trans. 1, 1998, 51.
21 H. Vorbru¨ggen, K. Krolikiewicz and B. Bennua, Chem. Ber., 1981, 114,
1234.
22 V. Nair and S. G. Richardson, J. Org. Chem., 1980, 45, 3969.
23 A. Al Mourabit, M. Beckmann, A. Poupat, A. Ahond and P. Potier,
Tetrahedron: Asymmetry, 1996, 7, 3455.
ring had been effectively deleted, was readily accessible by
condensation of 4 with N-trimethylsilylimidazole.23 A small
amount of bis-glycosylated material was also formed. The
desired product was deprotected, phosphorylated and depro-
tected as previously. We also synthesised an analogue (‘ur-
idophostin’, 21†) possessing the natural nucleic acid base,
uracil. Condensation of 4 with 4-methylanisole gave both a-
1
and b-substituted aryl C-glycosides in a ca. 1+1 ratio from H
NMR spectroscopy; these products were used to prepare the a-
and b-4-methyl anisole analogues of adenophostin A, 22a and
22b respectively, by the coupling of 4-methylanisole and 4 in
the presence of AgCO2CF3 and SnCl4. This demonstrates the
utility of our route also for the preparation of C-nucleoside
analogues.
Convergent construction of 2 was described by van Straten
et al.14 This elegant approach was somewhat hampered,
however, by a large number of protection/deprotection steps.
Our present route allows efficient access to a central disaccha-
ride 4, whilst the choice of protecting groups requires minimal
manipulation between Vorbru¨ggen condensation and target
trisphosphates.
In summary, we report here an efficient route to the first
synthetic base-modified analogues of adenophostin A.
We thank the Wellcome Trust for a Prize Studentship
(R. D. M.) and Programme Grant Support (045491 to
B. V. L. P.), the Ministry of Education, Science, Sports and
Culture of Japan for support to S. S. and acknowledge
discussions with Dr A. M. Riley and C. Mort for technical
assistance.
Communication a909347h
220
Chem. Commun., 2000, 219–220